Abstract: To address the challenge of predicting wake flow caused by the complex motion of floating wind turbines, this paper establishes a high-precision CFD numerical simulation model based on an improved actuator disk model. First, by developing a UDF program that accurately identifies body forces under multiple motion states, the paper systematically compares six simulation methods and validates them through wind tunnel tests. Then, we proposes an optimized solution combining the improved actuator disk model and the Realizable k-ε model, accurately capturing the velocity and turbulence intensity distribution characteristics of the turbine wake flow. Sensitivity analysis based on 80 single motion conditions reveals that the influence of incoming turbulence and thrust coefficient on the wake is significantly greater than that of motion parameters, while the impact of motion cycle is negligible. Among these, pitch, roll, and yaw are the dominant motion forms. Based on this, a Jensen-Gaussian-type wake model is proposed, with an average prediction error of only 2.0%, reducing errors by 3.8% to 6.6% compared to traditional Jensen and BPA wake models. For multiple coupled motion conditions, by comparing five wake super position models, the optimal coupled motion wake prediction method is established, achieving efficient and precise simulation under the synergistic effects of multiple motions. This study provides a high-precision, low-cost wake prediction toolchain for optimizing the layout and operational control of offshore wind farms in deep-sea environments.